There is increasing recognition that efforts to reduce disability among the elderly must be based on a better understanding of factors that slow the cascade from chronic illness to organ-level physical impairment to disability. Progression along this cascade is not inevitable, but may be buffered by personal and social resources, preventive adaptations, and impairment-relevant ameliorative adaptations. Arthritis is the most prevalent chronic medical condition for older people, and is one of the most common causes of physical impairment and disability. The factors that buffer the progression from arthritis-induced musculoskeletal impairments to personal and social disabilities have not been well studied. The propose study seeks to test a comprehensive model of buffers of disability and diminished quality of life for community-living elderly with and without arthritis. Our model also takes into account the important effects of comorbid conditions and potential gender differences in adapting to arthritis and attendant disability. The study sample consists of approximately 700 community-living old-old (age 75+) subjects who are enrolled in an ongoing NIA-funded MERIT-award study of Florida retirees. Fifty-four percent of the sample report having arthritis. Data collection for the proposed project will commence following four annual waves of the ongoing MERIT-award study. A three- wave longitudinal design will be used. The proposed study will take advantage of existing measures of chronic illness, personal (ADL/IADL) disability, health-promotion activities, social, psychological, and economic resources, and quality of life that we are gathering as part of our ongoing MERIT-award study. The proposed new study will introduce additional self-report data on arthritis, physical impairments, ameliorative adaptations, social disability and quality of life. Because our proposed study will be nested within the ongoing MERIT-award study, it can be feasibly performed within the budget constraints of the center grant proposal. We will use descriptive analyses to characterize the range of arthritis symptomatology and comorbid conditions and impairment, and their association with personal and social disability and diminished quality of life. We will also use standard uni-variate statistics and regression to test hypotheses about (10 the association of musculoskeletal impairment with specific disabilities, and (2) gender differences in effect of arthritis-related impairments on disability. Additionally, the proposed study will include multi-variate analyses of variance, least squares regression, and structural equations to test our comprehensive causal model regarding buffers of the disability cascade. More specifically, we will examine the buffering effect of preventive and ameliorative adaptations and general resources on the progression from arthritis to disability and diminished quality of life. Findings from the proposed study will be useful to clinicians, service providers, policy markers, and others attempting to minimize arthritis- induced disability and enhance quality of life among the elderly.